Skip to content

Consider publishing to pypi? #29

Description

@HandcartCactus

I was thinking about building something very similar myself, and I'm delighted to learn that I don't have to. The code looks well made and straightforward. Would you consider publishing this anytime soon? I'm happy to assist with any coding/tests/documentation.

  • Create setup.py
    • The distribution name metatune is taken on PyPI. Here are some options:
      • Ask ChrisCummins to unpublish their package (appears to be unused).
      • Add a prefix: python-metatune or sklearn-metatune
      • Choose a new name: e.g. sklearn-optuna (must be available on PyPI)
    • Select a version (0.0.1)
    • Select an author_name and email (@ches-001 I think these should be yours)
  • Actually Publish
    • Review setup.py
    • Discuss import name and distribution name
      • Consider changing import name to match distribution name, if metatune is not used.
    • Build project (Please let me know if you prefer a specific build back-end.)
    • Discuss adding tox/nox or other CI/CD tooling
    • Publish project on test.pypi.org and test import
    • Publish project on pypi.org

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions